SOURCE-LINKED INTELLIGENCE

Scaling Model-Generated Distillation Data Can Make Latent Teacher Traits More Recoverable

arXiv · AI, language, vision and robotics · article · Aug 27, 2026 · UTC

Scaling model-generated data is usually viewed as improving distillation: more examples should increase coverage, reduce noise, and produce stronger students. We show a second effect: larger datasets can make subtle teacher-specific signals easier to detect in the trained student, even when examples are off-task and never mention the trait. In a controlled setup inspired by subliminal learning, a teacher induced to express a target trait generates restricted off-task data, such as number-only completions. Students trained on different amounts of independent off-task data are evaluated in a sep

Read original source ↗ Open in workspace

recordType

paper

region

Global